提交 a5daf0e3 编写于 作者: V Vasily Gorbik 提交者: Martin Schwidefsky

s390/decompressor: swap .text and .rodata.compressed sections

.rodata.compressed section contains compressed linux image and is
quite large. By swapping text and rodata.compressed sections, the
decompressor code ends up between 0x11000 and 0x100000 addresses,
which makes it easier:
- to distinguish the decompressor phase from decompressed code (which
lives above 0x100000, except for small startup_continue),
- define break points which don't intersect with the main kernel image
later.
Signed-off-by: NVasily Gorbik <gor@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>
上级 401a0b8c
...@@ -17,15 +17,15 @@ SECTIONS ...@@ -17,15 +17,15 @@ SECTIONS
HEAD_TEXT HEAD_TEXT
_ehead = . ; _ehead = . ;
} }
.rodata.compressed : {
*(.rodata.compressed)
}
.text : { .text : {
_text = .; /* Text */ _text = .; /* Text */
*(.text) *(.text)
*(.text.*) *(.text.*)
_etext = . ; _etext = . ;
} }
.rodata.compressed : {
*(.rodata.compressed)
}
.rodata : { .rodata : {
_rodata = . ; _rodata = . ;
*(.rodata) /* read-only data */ *(.rodata) /* read-only data */
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册